reminder that you can create an AI video for literally ANYTHING the prompt is everything.. this is a result of our new V2 prompting system: https://t.co/2klauyXVFc

Miko

@Mho_23 · creator

TWITTER · 2026-03-07Source

107likes

16comments

Remix This

Prompt

GLOBAL LOCK: 
Subject is an elderly woman, approximately 80-85 years old, with long, silver-grey hair tied in a loose side braid. She has a warm, weathered complexion with natural wrinkles and a kind expression. She is wearing a cream-colored, textured linen shirt and a chunky turquoise bead necklace. The environment is a cozy, warm-toned podcast studio with a professional black Shure SM7B microphone on a boom arm in the immediate foreground. A ceramic mug sits on the wooden table. The background is softly blurred (bokeh) with warm orange floral accents and a macramé wall hanging. Lighting is soft, directional studio lighting creating a cinematic feel. The color grade is warm, earthy, and high-contrast. Audio is a warm, slightly raspy female voice with a rhythmic, storytelling cadence.

[00:00–00:04]
Subject: Close-up (MCU) of the elderly woman. She is looking slightly off-camera as if talking to an interviewer, smiling warmly.
Action: Her lips move in sync with the words "People always ask me how I look like this at eighty-two." She tilts her head slightly.
Camera: Static MCU, 50mm lens feel, shallow depth of field.
Speech: "People always ask me how I look like this at eighty-two." Warm, inviting tone. High lip-sync strictness.
Lighting: Soft key light from the left, gentle rim light on her hair.

[00:04–00:07]
Subject: Same MCU. She lets out a soft, genuine laugh, her eyes crinkling.
Action: She looks down and then back up, gesturing slightly with her head. "I mean, yeah, my grandmother taught me this when I was a little girl."
Camera: Static, maintaining focus on her facial expressions.
Speech: "I mean, yeah, my grandmother taught me this when I was a little girl." Conversational, nostalgic pace.
Motion: Natural hair movement as she laughs.

[00:07–00:10]
Subject: Same MCU. Her expression becomes more focused and intentional.
Action: She leans in slightly. "Sea moss. Every single morning." She emphasizes "Sea moss" with a knowing nod.
Camera: Static, but the slight lean creates a sense of intimacy.
Speech: "Sea moss. Every single morning." Clear enunciation on "Sea moss," followed by a short pause.
Lighting: Consistent warm studio glow.

[00:10–00:14]
Subject: Same MCU. She continues her explanation, looking wise and authoritative.
Action: "Ninety-two of the hundred and two minerals your body needs." Her facial muscles move naturally with the complex syllables.
Camera: Static MCU.
Speech: "Ninety-two of the hundred and two minerals your body needs." Informative, rhythmic delivery.
Motion: Subtle micro-movements in the background (slight blur of the floral elements) to maintain realism.

NEGATIVE PROMPT: 
Visual: robotic movement, plastic skin texture, flickering lights, inconsistent hair length, extra fingers, distorted microphone shape, blurry face, low resolution, jittery background, morphing clothes.
Speech: monotone voice, robotic cadence, misaligned lip-sync, harsh 's' sounds (sibilance), loud background hiss, unnatural pauses, popping 'p' sounds (plosives).

SPEECH PACK:
Transcript:
"People always ask me how I look like this at eighty-two. I mean, yeah, my grandmother taught me this when I was a little girl. Sea moss. Every single morning. Ninety-two of the hundred and two minerals your body needs."

TAKE_A (Storyteller): Warm, slow, emphasizing the "grandmother" and "sea moss" with a smile.
TAKE_B (Expert): Slightly faster, more authoritative on the "ninety-two of the hundred and two" part.
TAKE_C (Casual): More laughter in the middle, very relaxed and conversational.

Prosody Markup:
"People always ask me... [pause] how I look like this at **eighty-two**. [laugh] I mean, yeah... my grandmother taught me this when I was a **little girl**. [pause] **Sea moss**. Every. Single. Morning. **Ninety-two** of the hundred and two minerals your body needs."

How Mho_23 Made This Longevity Podcast Guest Video — and How to Recreate It

This case study analyzes a high-fidelity AI-generated video featuring an elderly woman sharing her longevity secrets in a cinematic podcast setting. The video leverages the "wise elder" archetype combined with the high-authority "podcast interview" aesthetic. Key visual elements include warm, directional studio lighting, a shallow depth of field (bokeh), and a highly consistent character model. The content taps into the evergreen health and wellness niche, specifically the "sea moss" trend, making it a perfect example of how to use AI to create authoritative, niche-specific content without a physical studio or talent.

What You’re Seeing

The video features a close-up (MCU) of an elderly woman with long, silver-grey hair tied in a loose braid. She wears a cream-colored linen shirt and a prominent turquoise bead necklace, suggesting a specific cultural or "naturalist" persona. She sits behind a professional Shure SM7B-style microphone on a boom arm, with a ceramic mug in the foreground creating a sense of depth. The background is a warm, blurred interior with hints of macramé and orange floral accents, mimicking a high-end podcast studio.

Shot-by-shot Breakdown

Time Range	Visual Content	Shot Language	Lighting & Tone	Viewer Intent
00:00–00:04	Subject introduces the "82-year-old" hook.	Medium Close-Up (MCU), static.	Warm key light, soft shadows.	Establish authority & curiosity.
00:04–00:07	Subject laughs, mentions her grandmother.	MCU, natural head tilt/movement.	Consistent warm palette.	Build relatability & heritage.
00:07–00:10	Reveal of the "secret": Sea Moss.	MCU, slight forward lean.	High clarity on facial details.	Deliver the "value" point.
00:10–00:14	Explanation of mineral benefits.	MCU, expressive hand/head motion.	Cinematic, earthy tones.	Reinforce expertise/logic.

Why It Went Viral

The success of this video lies in the "Authority-Contrast" mechanism. By placing an elderly woman—who naturally commands respect and curiosity regarding longevity—in a modern podcast setting, the creator creates instant credibility. The "82-year-old" hook is a powerful biological trigger; humans are evolutionarily wired to seek out the secrets of those who have successfully aged. The mention of "Sea Moss" connects the video to a trending health topic, ensuring it hits the right SEO and algorithm triggers for wellness-seeking audiences.

From a platform perspective, the video excels in retention and "save-ability." The dynamic text overlays (using a stylish serif font) keep the viewer's eyes moving and reduce the cognitive load of processing the speech. The high production value (lighting, microphone, bokeh) signals "quality content," which prevents users from scrolling past. The loop potential is high because the information is delivered in a punchy, "fact-dense" manner that viewers might want to watch twice to fully absorb the "mineral count" mentioned at the end.

5 Viral Hypotheses

The "Wise Elder" Hook: Using an older character to deliver health advice increases trust by 3x compared to a younger influencer. (Observation: The subject's age is the first thing mentioned).
Podcast Aesthetic = Authority: The presence of a professional microphone subconsciously tells the viewer "this person is worth interviewing." (Observation: The mic is a central prop).
Specific Numbers: Mentioning "92 of 102 minerals" provides a concrete "fact" that viewers are more likely to save for reference. (Observation: The specific stat at the end).
Cultural Visual Cues: The turquoise necklace and linen shirt suggest "natural/holistic" living, which aligns with the sea moss message. (Observation: Wardrobe choice).
Dynamic Typography: Using a mix of fonts and sizes for subtitles keeps the visual rhythm high even in a static shot. (Observation: The "People always" text style).

How to Recreate (Step-by-Step)

Niche Selection: Choose a high-trust niche like Health, Finance, or Philosophy where a "wise" character adds value.
Character Design (Midjourney): Generate a consistent character sheet. Use prompts like: elderly woman, silver hair, linen clothing, turquoise jewelry, hyper-realistic, 8k --ar 16:9.
Environment Setup: Create a background image of a "warm podcast studio" with a professional microphone and shallow depth of field.
Scripting: Write a 15-second script with a strong "I'm [Age] and here is my secret" hook.
Voice Generation (ElevenLabs): Use a "warm, raspy, elderly" voice profile. Add slight pauses for "grandmotherly" pacing.
Video Generation (Kling/Luma/Runway): Use the character image + voice as input. Focus on "lip-sync" and "natural head movement" settings.
Editing & Captions: Use CapCut or Premiere. Add serif-style dynamic captions. Ensure the text "pops" on the screen.
Sound Design: Add a very low-volume, warm "lo-fi" or "ambient studio" background track to fill the silence between words.

Growth Playbook

Opening Hooks

"I'm 82 years old, and this is the one thing I never skip..."
"My grandmother lived to 105. This was her daily ritual."
"Stop buying expensive supplements. You only need this one plant."

Caption Templates

The "Secret" Template:
People always ask how I keep my energy up at [Age]. 🌿 It’s not luck, it’s a ritual passed down through generations. Have you tried [Product] yet? 👇 #longevity #healthtips

The "Science" Template:
Did you know your body needs [Number] minerals to function? 🧬 Most of us are missing out. Here is how I get mine every single morning. Save this for your next grocery trip! 🛒 #wellness #biohacking

Hashtag Strategy

Broad: #health #wellness #longevity #aiart
Mid-tier: #seamossbenefits #holistichealth #podcastclips
Niche: #aiinfluencer #aginggracefully #naturalremedies

FAQ

What tools make it look the most similar?

Use Midjourney for the base image and Kling AI or Luma Dream Machine for the most realistic lip-sync and movement.

What are the 3 most important words in the prompt?

"Cinematic," "Subsurface scattering" (for skin), and "Shallow depth of field."

Why does the generated face look inconsistent?

You need to use a "Character Reference" (cref) in Midjourney or a consistent seed in your video generator.

How can I avoid making it look like AI?

Add "micro-movements" like blinking and slight head tilts, and ensure the lighting on the subject matches the background.

Is it easier to go viral on Instagram or TikTok with this?

Instagram Reels tends to favor this "high-aesthetic/authoritative" podcast style more than TikTok's raw UGC vibe.